Skip to main content
Blog
geminigemini 4gemini-4-hydrogenpricing

Gemini 4 Hydrogen API Pricing and Day 0 Support on LiteLLM

Gemini 4 Hydrogen on LiteLLM: the LiteLLM and Gemini 4 Hydrogen lockup, then Gemini 4 Hydrogen selected in the LiteLLM model list next to Gemini 4 Carbon, Gemini 4 Argon and Gemini 3.8 Flash.
Published:
Misbah SyedDevRel Engineer, LiteLLMMateo WangAI Engineer, LiteLLMKerry LuSoftware Engineer, LiteLLM

LiteLLM now supports Gemini 4 Hydrogen on Day 0. Google built Hydrogen as the fast, everyday model for coding and agentic workflows, and it scores 51 on the Artificial Analysis Intelligence Index against Gemini 4 Argon's 50. Use it on Google AI Studio and Gemini Enterprise Agent Platform (formerly Vertex AI) through the LiteLLM AI Gateway, with spend, rate limits and logs in one place.

Gemini 4 Hydrogen pricing​

Prices per 1M tokens, the same across the full 1M-token context window:

  • Input: $1.50 / MTok
  • Output: $7.50 / MTok. Thinking tokens bill as output.
  • Cache read: $0.075 / MTok. Hydrogen supports implicit caching only.
  • Batch and Flex: $0.75 input and $3.75 output
  • Priority: $2.70 input and $13.50 output

Gemini 4 Argon and Gemini 4 Carbon cost $2.00 / $10.00 until November 30 and $4.00 / $20.00 from December 1, so Hydrogen is the lowest-priced Gemini 4 model. On Gemini Enterprise Agent Platform, regional endpoints cost 10% more than global.

Details on pricing: https://models.litellm.ai/models/gemini-4-hydrogen

Use Gemini 4 Hydrogen with LiteLLM​

No upgrade needed: open Price Data under Models + Endpoints in the UI and click Reload Price Data (or POST /reload/model_cost_map), on v1.76.0 and above.

1. Setup config.yaml

model_list:
  - model_name: gemini-4-hydrogen
    litellm_params:
      model: gemini/gemini-4-hydrogen
      api_key: os.environ/GEMINI_API_KEY

2. Start the proxy

docker run -d \
  -p 4000:4000 \
  -e GEMINI_API_KEY=$GEMINI_API_KEY \
  -v $(pwd)/config.yaml:/app/config.yaml \
  ghcr.io/berriai/litellm:v1.106.2 \
  --config /app/config.yaml

3. Test it!

curl --location 'http://0.0.0.0:4000/chat/completions' \
--header 'Content-Type: application/json' \
--header "Authorization: Bearer $LITELLM_KEY" \
--data '{
  "model": "gemini-4-hydrogen",
  "messages": [
    {
      "role": "user",
      "content": "what llm are you"
    }
  ],
  "reasoning_effort": "medium"
}'

LiteLLM sends reasoning_effort to Hydrogen as its thinking_level.

FAQ​

How much does Gemini 4 Hydrogen cost?​

$1.50 per 1M input tokens and $7.50 per 1M output tokens, with no surcharge for long prompts. Batch and Flex halve both prices.

Is Gemini 4 Hydrogen cheaper than Gemini 4 Argon?​

Yes. Hydrogen costs $1.50 / $7.50 per 1M tokens. Argon and Carbon cost $2.00 / $10.00 until November 30 and $4.00 / $20.00 from December 1.

Is Gemini 4 Hydrogen free on Google AI Studio?​

Yes. Google AI Studio lists a free tier for Hydrogen on the Standard tab. Google Search grounding is paid tier only.

What is the context window of Gemini 4 Hydrogen?​

1,048,576 tokens for input, thinking and output together. Google returns up to 65,536 output tokens per synchronous request, and LiteLLM continues past that cap the same way it does for Argon.

Which thinking levels does Gemini 4 Hydrogen support?​

low, medium, high and xhigh, with high as the default. Set low, medium or high with reasoning_effort. Support for xhigh through reasoning_effort is tracked in PR #47086.

How do I switch from Gemini 3.8 Flash to Gemini 4 Hydrogen?​

Change gemini/gemini-3.8-flash to gemini/gemini-4-hydrogen in your config. Google says Hydrogen finishes tasks 2.5x faster than 3.8 Flash.

Does Gemini 4 Hydrogen support context caching?​

Implicit caching only, from 4,096 prompt tokens. Explicit caching is not supported.

Is Google Search grounding free on Gemini 4 Hydrogen?​

The first 5,000 queries a month are free, shared with Gemini 3 and newer models, then $14 per 1,000 queries. Grounding with Google Maps has the same allowance and price.

How do I call Gemini 4 Hydrogen with an OpenAI-compatible API?​

Add gemini/gemini-4-hydrogen to your LiteLLM config. Then send a standard /chat/completions request to the proxy with "model": "gemini-4-hydrogen".

Feedback​

Questions and feedback go in GitHub discussion #51436.

References​